From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756370Ab0GHPAh (ORCPT ); Thu, 8 Jul 2010 11:00:37 -0400 Received: from mail.openrapids.net ([64.15.138.104]:33734 "EHLO blackscsi.openrapids.net" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751749Ab0GHPAg (ORCPT ); Thu, 8 Jul 2010 11:00:36 -0400 Date: Thu, 8 Jul 2010 11:00:33 -0400 From: Mathieu Desnoyers To: Ingo Molnar Cc: "Ma, Ling" , "linux-kernel@vger.kernel.org" , "H. Peter Anvin" Subject: Re: [PATCH -tip] Revert x86, mem: Optimize memcpy by avoiding memory false dependece Message-ID: <20100708150033.GA10224@Krystal> References: <20100708062733.GA24740@Krystal> <20100708071153.GH4414@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100708071153.GH4414@elte.hu> X-Editor: vi X-Info: http://www.efficios.com X-Operating-System: Linux/2.6.26-2-686 (i686) X-Uptime: 11:00:21 up 166 days, 17:37, 7 users, load average: 0.04, 0.05, 0.00 User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Ingo Molnar (mingo@elte.hu) wrote: > > * Mathieu Desnoyers wrote: > > > * Ma, Ling (ling.ma@intel.com) wrote: > > > Hi Mathieu > > > > > > I did git revert as bellow > > > Revert "x86, mem: Optimize memcpy by avoiding memory false dependece" > > > > > > This reverts commit 77ff5e49c20bb72a99a12bfb00dbf90b2302f087. > > > And I found the machine will crash with your config file > > > > > > Could you revert the patch : 77ff5e49c20bb72a99a12bfb00dbf90b2302f087 on your > > > machine, and check it again ? > > > > Hi Ling, > > > > I just successfully booted my machine 3 times in a row with a revert of > > commit a1e5278e40f16a4611264f8da9e557c16cb6f6ed from -tip. This is actually > > the same as reverting commit 77ff5e49c20bb72a99a12bfb00dbf90b2302f087, as > > a1e5278e40f16a4611264f8da9e557c16cb6f6ed is simply a merge of that commit > > (the diffs are exactly the same). > > > > Without this revert, the machine gets to: > > Please check latest -tip (2a74852 or later), the commit should be gone now. Will do, thanks! Mathieu > > There's clearly some (boundary-condition?) bug in the assembly code of > 77ff5e49c2. > > Thanks, > > Ingo -- Mathieu Desnoyers Operating System Efficiency R&D Consultant EfficiOS Inc. http://www.efficios.com